It seems I misunderstood Sunil's response and somewhat freaked out
because it appeared that he was giving the wrong method for making a
QQ plot, but was actually demonstrating the sampling variability. My
apologies to Sunil.
2009/9/27 Duncan Murdoch :
> Eric Thompson wrote:
>>
&
otting position (must be careful here in general!)
p <- ppoints(n)
# Compute the quantile
x.q <- qnorm(p)
points(x.q, x.s, col = "red")
# and they fall exactly on the points generated by qqnorm().
Now, you should be able to generalize this for any distribution. Hope
this helps.
E
it would be
desirable for frequency to be defined over the same range as in
spec.pgram. All that would need to be added would be a line to scale
the freq vector using the sampling frequency before it is returned.
Eric Thompson
Graduate Student
Dept. of Civil & Env. Eng.
Tufts University
&
There are probably much more fancy ways to do this, but since no one
else has posted a response, here's what I would do:
> plot(seq(10, 40, 10), axes = FALSE)
> axis(1, at = 1:4, lab = LETTERS[1:4])
> axis(side = 2)
> box()
On 10/20/07, Yong Wang <[EMAIL PROTECTED]> wrote:
> Dear R-list
>
> My q
Try
par(mar=c(3,4,2,2), mfrow=c(5,2))
On 10/12/07, Leeds, Mark (IED) <[EMAIL PROTECTED]> wrote:
> I am constructing plots ( regular not lattice ) and my initial command
> is
>
> par(mar=c(3,4,2,2), mfcol=c(5,2))
>
> and then I create 10 plots on the page. It looks great but the plots on
> the pa
Have you tried using pdf(), postscript(), or jpeg()? All of these have
arguments to specify the height and width of the device.
On 10/8/07, Christoph Krammer <[EMAIL PROTECTED]> wrote:
> Hello,
>
> I have another (possibly easy) question:
>
> How to specify the size of a plot? When I draw a plot
...) :
> need finite 'xlim' values
> In addition: Warning messages:
> 1: NAs introduced by coercion in: as.double.default(x)
> 2: no non-missing arguments to min; returning Inf in: min(x)
> 3: no non-missing arguments to max; returning -Inf in: max(x)
> 4: NAs in
TED]> wrote:
> Thanks for your quick reply, Eric.
>
> I want plot colnames(n) as string on x-axis. If the regression lines don't
> fit the data very well, it is OK, the plot is only for quality check.
>
>
> On Tue, 2 Oct 2007, Eric Thompson wrote:
>
> If I've cor
If I've correctly interpreted what you want, you first need to get the x values:
x <- colnames(n)
x <- as.numeric(substr(x, 1, nchar(x) - 1))
Then it seems fairly easy to use matplot to get the values with
different colors for each concentration
dim(x) <- c(length(x), 1)
matplot(x, t(n), pch = 1
Have you tried filled.contour()? It automatically generates a legend.
However, I'm not sure what is going on with your x's and y's being
sorted random numbers. I assume your actual data are not like this. If
you do not have data in an equally spaced grid you may need to create
one by some sort of i
Did you try reading the help pages or looking at the examples in the
help pages? Did you even make an attempt at using the command? If so,
did you get an error or unexpected results?
I don't know how you can expect people to help you when you appear to
have made no effort yourself.
On 10/1/07,
11 matches
Mail list logo
